Whisper Large V3 Persian Common Voice 17
A Persian automatic speech recognition model fine-tuned based on Whisper Large v3, trained using the Common Voice 17 dataset, significantly improving Persian recognition accuracy.
Downloads 442
Release Time : 3/15/2025
Model Overview
This is an automatic speech recognition model specifically optimized for Persian, based on OpenAI's Whisper Large v3 architecture, fine-tuned on the Persian subset of Mozilla Common Voice 17.
Model Features
Large-scale data training
Trained with over 250,000 Persian speech samples, significantly improving recognition accuracy compared to previous versions (83,000 samples)
Low word error rate
Achieved a word error rate (WER) of 21.43 in Persian speech recognition
Specialized optimization
Specifically optimized for Persian language characteristics, improving recognition accuracy and robustness for this language
Model Capabilities
Persian speech recognition
Long audio processing (supports 30-second chunks)
Use Cases
Speech-to-text
Persian meeting transcription
Automatically convert Persian meeting recordings into text transcripts
Improved accuracy, reduced word error rate
Persian media subtitle generation
Automatically generate subtitles for Persian video content
Increased subtitle production efficiency
Featured Recommended AI Models
Š 2025AIbase